ICML 2024 Workshop HiLD Submissions

How Do Nonlinear Transformers Acquire Generalization-Guaranteed CoT Ability?
Hongkang Li, Meng Wang, Songtao Lu, Xiaodong Cui, Pin-Yu Chen
- Published: 16 Jun 2024, Last Modified: 18 Jun 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Learning Multi-Index Models with Neural Networks via Mean-Field Langevin Dynamics
Alireza Mousavi-Hosseini, Denny Wu, Murat A Erdogdu
- Published: 16 Jun 2024, Last Modified: 20 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Analysing feature learning of gradient descent using periodic functions
Jaehui Hwang, Taeyoung Kim, Hongseok Yang
- Published: 16 Jun 2024, Last Modified: 15 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Expressivity of Neural Networks with Fixed Weights and Learned Biases
Ezekiel Williams, Avery Hee-Woon Ryoo, Thomas Jiralerspong, Alexandre Payeur, Matthew G Perich, Luca Mazzucato, Guillaume Lajoie
- Published: 16 Jun 2024, Last Modified: 10 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Linear Weight Interpolation Leads to Transient Performance Gains
Gaurav Iyer, Gintare Karolina Dziugaite, David Rolnick
- Published: 16 Jun 2024, Last Modified: 15 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
The optimization landscape of Spectral neural network
Chenghui Li, Rishi Sonthalia, Nicolas Garcia Trillos
- Published: 16 Jun 2024, Last Modified: 16 Jun 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Do Parameters Reveal More than Loss for Membership Inference?
Anshuman Suri, Xiao Zhang, David Evans
- Published: 16 Jun 2024, Last Modified: 19 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
u-μP: The Unit-Scaled Maximal Update Parametrization
Charlie Blake, Constantin Eichenberg, Josef Dean, Lukas Balles, Luke Yuri Prince, Björn Deiseroth, Andres Felipe Cruz-Salinas, Carlo Luschi, Samuel Weinbach, Douglas Orr
- Published: 16 Jun 2024, Last Modified: 16 Jun 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Toward Global Convergence of Gradient EM for Over-Parameterized Gaussian Mixture Models
Weihang Xu, Maryam Fazel, Simon Shaolei Du
- Published: 16 Jun 2024, Last Modified: 19 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Merging Text Transformer Models from Different Initializations
Neha Verma, Maha Elbayad
- Published: 16 Jun 2024, Last Modified: 17 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Deep Networks Always Grok and Here is Why
Ahmed Imtiaz Humayun, Randall Balestriero, Richard Baraniuk
- Published: 16 Jun 2024, Last Modified: 17 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Nonconvex Meta-optimization for Deep Learning
Xinyi Chen, Evan Dogariu, Zhou Lu, Elad Hazan
- Published: 16 Jun 2024, Last Modified: 23 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Loss landscape geometry reveals stagewise development of transformers
George Wang, Matthew Farrugia-Roberts, Jesse Hoogland, Liam Carroll, Susan Wei, Daniel Murfet
- Published: 16 Jun 2024, Last Modified: 15 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
All Roads Lead to Rome? Exploring Representational Similarities Between Latent Spaces of Generative Image Models
Charumathi Badrinath, Usha Bhalla, Alex Oesterling, Suraj Srinivas, Himabindu Lakkaraju
- Published: 16 Jun 2024, Last Modified: 18 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
The Hidden Pitfalls of the Cosine Similarity Loss
Andrew Draganov, Sharvaree Vadgama, Erik J Bekkers
- Published: 16 Jun 2024, Last Modified: 19 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
The Empirical Impact of Neural Parameter Symmetries, or Lack Thereof
Derek Lim, Theo Putterman, Robin Walters, Haggai Maron, Stefanie Jegelka
- Published: 16 Jun 2024, Last Modified: 19 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Interpolated-MLPs: Controllable Inductive Bias
Sean Wu, Jordan Hong, keybai, Gregor Bachmann
- Published: 16 Jun 2024, Last Modified: 19 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Provable Benefit of Cutout and CutMix for Feature Learning
Junsoo Oh, Chulhee Yun
- Published: 16 Jun 2024, Last Modified: 20 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Fundamental limits of weak learnability in high-dimensional multi-index models
Emanuele Troiani, Yatin Dandi, Leonardo Defilippis, Lenka Zdeborova, Bruno Loureiro, Florent Krzakala
- Published: 16 Jun 2024, Last Modified: 16 Jun 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
A Unified Approach to Feature Learning in Bayesian Neural Networks
Noa Rubin, Zohar Ringel, Inbar Seroussi, Moritz Helias
- Published: 16 Jun 2024, Last Modified: 18 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Correlated Noise in Epoch-Based Stochastic Gradient Descent: Implications for Weight Variances
Marcel Kühn, Bernd Rosenow
- Published: 16 Jun 2024, Last Modified: 15 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Latent functional maps
Marco Fumero, Marco Pegoraro, Valentino Maiorca, Francesco Locatello, Emanuele Rodolà
- Published: 16 Jun 2024, Last Modified: 16 Jun 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Where Do Large Learning Rates Lead Us? A Feature Learning Perspective
Ildus Sadrtdinov, Maxim Kodryan, Eduard Pokonechny, Ekaterina Lobacheva, Dmitry Vetrov
- Published: 16 Jun 2024, Last Modified: 20 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
How Do Transformers Fill in the Blanks? A Case Study on Matrix Completion
Pulkit Gopalani, Ekdeep Singh Lubana, Wei Hu
- Published: 16 Jun 2024, Last Modified: 19 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone
Repetita Iuvant: Data Repetition Allows SGD to Learn High-Dimensional Multi-Index Functions
Luca Arnaboldi, Yatin Dandi, Florent Krzakala, Luca Pesce, Ludovic Stephan
- Published: 16 Jun 2024, Last Modified: 18 Jul 2024
- HiLD at ICML 2024 Poster
- Readers: Everyone